Quantum state data transmission method and apparatus, device, medium, and quantum simulation system

By constructing a shared storage pool and storage network in the quantum simulation system, the problem of limited data throughput in distributed scaling is solved, achieving efficient quantum simulation computing and improving the system's scalability and performance.

WO2026238061A1PCT designated stage Publication Date: 2026-11-19BEIJING SURSEN NETWORK TECH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
PCT/CN2026/096992
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2025-05-15
Filing Date
2026-05-14
Publication Date
2026-11-19

AI Technical Summary

Technical Problem

Quantum simulation systems suffer from limited data throughput and increased network latency during distributed scaling, leading to a decrease in overall simulation efficiency.

Method used

A shared storage pool is constructed for all servers, and a data transmission channel is established between each server and the shared storage pool through a storage network, so that any server can directly access the storage media in the shared storage pool, reducing horizontal data transmission.

Benefits of technology

It effectively solves the bottleneck problem of distributed expansion of quantum simulation systems, improves the distributed expansion capability and overall performance of the system, and meets the data transmission requirements of large-scale quantum simulation systems.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN2026096992_19112026_PF_FP_ABST
    Figure CN2026096992_19112026_PF_FP_ABST
Patent Text Reader

Abstract

The present application discloses a quantum state data transmission method and apparatus, a device, a medium, and a quantum simulation system. The method comprises: reading, from a shared storage pool via a storage network, partial quantum state data required by a server for executing a quantum computing simulation task; the server executing the quantum computing simulation task on the basis of the partial quantum state data; and storing, into the shared storage pool, quantum state data generated upon the execution of the quantum computing simulation task. In the method, the quantum state data is stored in the shared storage pool independent of servers, rather than in a private storage medium of each server, so that any server can directly access any storage medium via the storage network without relying on other servers, eliminating east-west data traffic in a distributed quantum simulation cluster, thereby effectively solving the principal bottleneck problem in distributed scaling of a quantum simulation system, improving the distributed scalability of the quantum simulation system, so as to meet the data transmission requirements of a large-scale quantum simulation system.
Need to check novelty before this filing date? Find Prior Art

Description

Quantum state data transmission methods, devices, equipment, media, and quantum simulation systems

[0001] This application claims priority to Chinese Patent Application No. 202510630394.7, filed on May 15, 2025, entitled "Method, Apparatus, Device, Medium and Quantum Simulation System for Quantum State Data Transmission", the entire contents of which are incorporated herein by reference. Technical Field

[0002] This application relates to the field of quantum simulation, specifically to a quantum state data transmission method, apparatus, device, medium, and quantum simulation system. Background Technology

[0003] Quantum computing leverages the properties of quantum mechanics to exponentially increase computing power, thereby solving complex computational problems. Based on this, quantum technology is developing rapidly, and quantum computing has attracted widespread attention in both academia and industry.

[0004] Quantum simulation uses classical computers to simulate quantum computing and is an essential tool for developing and debugging quantum algorithms. However, because quantum simulation requires storing and manipulating all possible quantum state data, the scale of this data grows exponentially with the number of qubits, creating an urgent need for distributed scaling. Quantum simulation is characterized by high-frequency ergodicity, requiring the transmission of large-scale quantum state data. Current distributed architectures store data on the individual storage media of each server, necessitating the transfer of large amounts of data between different servers during quantum simulation calculations. This limits data throughput, creating a bottleneck and impacting the overall efficiency of the quantum simulation system. Therefore, addressing the bottleneck problem of distributed scaling in quantum simulation systems is a pressing issue that needs to be resolved. Summary of the Invention

[0005] In view of this, embodiments of this application provide a quantum state data transmission method, apparatus, device, medium, and quantum simulation system, which improves the distributed scalability of the quantum simulation system by constructing a shared storage pool for each server.

[0006] In a first aspect, embodiments of this application provide a quantum state data transmission method applied to a quantum simulation system. The quantum simulation system includes at least two servers, at least one storage medium, and a storage network. The at least one storage medium constitutes a shared storage pool for storing quantum state data. Either of the at least two servers can access any storage medium in the shared storage pool via the storage network without traversing other servers. The method includes: reading partial quantum state data required by the server to simulate a quantum computing task from the shared storage pool via the storage network; performing a simulated quantum computing task based on the partial quantum state data; and storing the quantum state data after performing the simulated quantum computing task in the shared storage pool.

[0007] In one possible implementation, at least one storage medium is located in at least one storage device independent of at least two servers, the at least one storage device being separate from the at least two servers.

[0008] In one possible implementation, the shared storage pool is mounted as a block device on at least two servers.

[0009] In one possible implementation, the block devices are shared block devices. Alternatively, all or some of the block devices are shared block devices, which are shared by multiple servers.

[0010] In one possible implementation, reading partial quantum state data required by the server to simulate a quantum computing task from a shared storage pool via a storage network includes: the server reading partial or all of the quantum state data from the shared storage pool via the storage network; and storing partial or all of the quantum state data in the server's cache. Optionally, the server's cache can be the server's memory.

[0011] In one possible implementation, the interface protocol of the storage medium is the same as the protocol used by the storage network during the data reading process of reading the part of the quantum state data required for the simulated quantum computing task from the shared storage pool via the storage network, and during the data storage process of storing the quantum state data after performing the simulated quantum computing task to the shared storage pool.

[0012] In one possible implementation, the storage network is any one or a combination of CXL, PCIe, SAS, IB, RoCE, and Ethernet networks.

[0013] In one possible implementation, at least one storage medium is one or a combination of high-bandwidth memory (HBM), phase-change memory (PCM), resistive random access memory (RRAM), static random access memory (SRAM), dynamic random access memory (DRAM), magnetoresistive random access memory (MRAM), flash memory, 3D XPoint, and hard disk drive (HDD).

[0014] In one possible implementation, the simulated quantum computing task is to simulate at least one quantum gate operating on quantum state data, wherein the simulated quantum computing task performs a specified simulated quantum computing operation on a specified range of quantum state data.

[0015] In one possible implementation, simulating the operation of at least one quantum gate on quantum state data includes: determining a set of simulated quantum computing tasks for each of at least two servers, based on the type and parameters of at least one quantum gate, wherein the set of simulated quantum computing tasks consists of zero or at least one simulated quantum computing task.

[0016] In one possible implementation, the server simulates quantum state data within a corresponding range in the shared memory pool by: determining a simulation computation task plan for each of at least two servers based on a set of quantum gate types and parameters, wherein the simulation computation task plan determines the simulation computation tasks that the server needs to execute; when executing each simulation computation task, each server determines the range of quantum state data in the shared memory pool corresponding to the task, reads the corresponding quantum state data, performs the corresponding calculations according to the requirements of the task, and finally updates the quantum state data in the shared memory pool according to the calculation results.

[0017] In one possible implementation, the simulation computation task plan includes zero or at least one simulation computation task. In an extreme case, it is possible that when simulating a quantum gate, only some servers need to perform the computation, while others remain idle and waiting; in this case, the simulation computation task plan for that server is empty.

[0018] Secondly, embodiments of this application provide a quantum simulation system, which includes: at least two servers, at least one storage medium, and a storage network, wherein the at least one storage medium constitutes a shared storage pool for storing quantum state data, any one of the at least two servers accesses any storage medium in the shared storage pool through the storage network, and the at least two servers are used to execute the quantum state data transmission method of the first aspect.

[0019] Thirdly, embodiments of this application provide a quantum state data transmission device applied to a quantum simulation system. The quantum simulation system includes at least two servers, at least one storage medium, and a storage network. The at least one storage medium constitutes a shared storage pool for storing quantum state data. Either of the at least two servers can access any storage medium in the shared storage pool via the storage network without going through other servers. The device includes: a data reading module configured to read partial quantum state data required by the server to perform a simulated quantum computing task from the shared storage pool via the storage network; a task execution module configured to execute a simulated quantum computing task based on the partial quantum state data; and a data storage module configured to store the quantum state data generated after executing the simulated quantum computing task to the shared storage pool.

[0020] Fourthly, embodiments of this application provide an electronic device including at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the quantum state data transmission method described in the first aspect.

[0021] Fifthly, embodiments of this application provide a computer-readable storage medium storing instructions that, when executed by a processor of an electronic device, enable the electronic device to perform the quantum state data transmission method described in the first aspect.

[0022] This application provides a quantum state data transmission method, apparatus, device, medium, and quantum simulation system. The method constructs at least one storage medium as a shared storage pool and establishes a data transmission channel between each server and the shared storage pool through a storage network. This means that quantum state data is stored in a shared storage pool independent of each server, rather than in a private storage medium for each server. This allows any server to directly access any storage medium in the shared storage pool through the storage network without relying on other servers. Throughout the quantum simulation calculation process, no horizontal quantum state data transmission is required between different servers, eliminating east-west data flow in the distributed quantum simulation cluster. This effectively solves the biggest bottleneck problem in the distributed expansion of quantum simulation systems, improves the distributed expansion capability of quantum simulation systems, and further enhances the overall performance of the system to meet the data transmission requirements of large-scale quantum simulation systems. Attached Figure Description

[0023] The accompanying drawings are provided to further illustrate the present application and form part of the specification. They are used together with the embodiments of the present application to explain the application and do not constitute a limitation thereof. The above and other features and advantages will become more apparent to those skilled in the art from the detailed example embodiments described with reference to the accompanying drawings, in which:

[0024] Figure 1 is a schematic diagram of the structure of a quantum simulation system provided in some embodiments of this application.

[0025] Figure 2 is an exemplary flowchart of a quantum state data transmission method provided in some embodiments of this application.

[0026] Figure 3 is a schematic diagram showing the correspondence between the range of quantum state data simulated by the server and the shared storage pool provided in some embodiments of this application.

[0027] Figure 4 is a schematic diagram of a quantum state data transmission device provided in some embodiments of this application.

[0028] Figure 5 is a schematic diagram of the structure of an electronic device provided in some embodiments of this application. Detailed Implementation

[0029] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0030] Application Overview

[0031] Because quantum simulation systems need to store and manipulate all possible quantum state data during the simulation of quantum computing, the demand for storage and data throughput increases exponentially as the number of qubits to be simulated increases.

[0032] Taking a full-amplitude quantum simulation system as an example, such a system achieves quantum computing by simulating the full amplitude of quantum states. Executing quantum computing operations requires storing and manipulating all possible quantum state data. For each additional qubit, the data space doubles. Furthermore, each quantum gate involved in the computation needs to traverse the data space. Due to memory limitations, a single server can only simulate a limited number of qubits; a full-amplitude quantum simulation typically does not exceed 34 qubits, resulting in approximately 256 GB of quantum state data. Therefore, for large-scale quantum simulation systems, the scale of quantum state data to be processed grows exponentially, leading to a situation where a single server cannot meet the data space requirements of the simulation system. Related technologies address this issue by using a traditional distributed architecture in quantum simulation systems. However, if a traditional distributed structure is used for quantum state data transmission, the high-frequency and large-scale data flow during the quantum simulation process puts severe pressure on the system's lateral bandwidth, limiting data throughput, increasing network latency, and ultimately affecting system performance.

[0033] Taking a qubit count of n=40 as an example, corresponding to 240 complex numbers, each occupying 16 bytes, the total data volume is 240 × 16 bytes (approximately 17592 GB). In a typical correlated distributed network, the transmission rate is 1.25 GB / s. Assuming the network rate is increased tenfold to 12.5 GB / s, transmitting the entire quantum state data would still take 1407 seconds. For a simple single-qubit operation (such as a Hadamard gate), addition and multiplication operations are required for each complex number. Each operation includes two additions and two multiplications, totaling 4 floating-point operations. Based on this, the total number of floating-point operations is 4 × 2n, meaning that for n=40, the computational load is 4 × 240 floating-point operations. High-performance GPUs can achieve a computing power on the order of 10¹⁵ FLOPS, which means they can complete 227 quantum gate operations per second. Compared to the GPU's ability to process 227 quantum gates per second, the current TCP / IP transmission network can only support data transmission of approximately 1 / 1407 ≈ 0.00071 quantum gates. In other words, the data throughput of the traditional distributed network architecture is several orders of magnitude lower than the computing power, which is far from meeting the demand for computing power. This forms a bottleneck for the distributed expansion of quantum simulation systems and significantly limits the overall performance of quantum simulators.

[0034] In related technologies, the architecture based on TCP / IP networks is optimized through GPU interconnection to solve the problems of high latency and limited throughput. However, this solution can only improve data throughput to a certain extent, and the improvement is limited. It does not solve the problem at its root. In other words, quantum simulation is characterized by high-frequency traversal of massive amounts of data, and existing distributed quantum simulation cluster technologies have serious east-west traffic bottlenecks, making them unsuitable.

[0035] To address the aforementioned issues, this application's embodiments, targeting the high data throughput requirements of large-scale quantum simulation systems, creatively construct a shared storage pool for each server and establish a data transmission channel between each server and the shared storage pool through a storage network. This innovatively changes the storage of quantum state data from private storage pools within each server to a shared storage pool, and establishes a data transmission channel between each server and each storage medium within the shared storage pool through the storage network. This allows any server to directly access any storage medium without relying on other servers. In other words, the technical solution provided by this application's embodiments eliminates east-west traffic at the cost of slightly extending the north-south data path, fundamentally solving the biggest bottleneck problem in the distributed scaling of quantum simulation computing and improving the distributed scaling capability of quantum simulation systems.

[0036] For example, in this embodiment of the application, a quantum algorithm can be represented by a quantum circuit composed of several quantum gates. These quantum gates are divided into several groups, and the simulation computation of each group of quantum gates is divided into several computational tasks. Each computational task involves a designated server performing a computation on quantum state data within a specified range using a specified algorithm. When executing each simulated quantum computation task, each server can read the corresponding quantum state data from a shared memory pool, execute the specified computational algorithm, and update the quantum state data stored in the shared memory pool based on the computation results.

[0037] Generally, the range of quantum state data written back is the same as the range of quantum state data read. However, depending on the type of quantum gate or the optimization and improvement of the simulation algorithm, it is possible that only a portion of the data within the read range needs to be updated, or even a different portion of the data needs to be updated. In other words, when performing a simulated quantum computing task, the range of data read and the range of data stored after computation may be the same, the latter may be a subset of the former, a portion of the latter may exceed the range of the former, or the latter may not overlap with the former at all.

[0038] Various non-limiting embodiments of this application will now be described in detail with reference to the accompanying drawings.

[0039] Exemplary quantum simulation system

[0040] This application provides a quantum simulation system 100. Figure 1 is a schematic diagram of the shared architecture of the quantum simulation system provided in some embodiments of this application. As shown in Figure 1, the quantum simulation system 100 may include at least two servers 111, at least one storage medium, and a storage network 112. The at least two servers 111 and at least one storage medium are connected through the storage network 112. The at least one storage medium constitutes a shared storage pool, and the at least two servers 111 can access any storage medium in the shared storage pool through the storage network 112.

[0041] In some embodiments, the quantum simulation system 100 can be a multi-network zone quantum simulation system. Specifically, the multi-network zone quantum simulation system can include at least two network zones 110 and a storage network 112. The storage network 112 can include a storage network access network 1121 and a storage network aggregation network 1122. Each network zone 110 can include at least one storage medium, a storage network access network 1121, and an optional server 111. The optional server 111 in the same network zone is connected to at least one storage medium through the storage network access network 1121. The access networks 1121 of different network zones 110 are also interconnected through the storage network aggregation network 1122, forming a unified large storage network. The storage media of each network zone constitute a unified large shared storage pool.

[0042] In some embodiments, the quantum simulation system can be composed of a quantum simulation system with multiple storage zones. Each server in a storage zone can read and write every storage medium in that storage zone through the storage network, but may not be able to read and write every storage medium in other storage zones.

[0043] As shown in Figure 1, in some embodiments, at least one storage medium may be located in at least one storage device (e.g., JBOD / JBOF / e-JBOF) 113, which is independent of the at least two servers.

[0044] In some embodiments, the shared storage pool may consist of all storage media in at least one storage device, i.e., all storage media are shared; or it may consist of some storage media in at least one storage device, i.e., some storage media are shared; or all or part of the space of all or part of the storage media may be shared. The size of the shared storage media can be flexibly set according to the scale of the quantum simulation system, without specific limitations.

[0045] In some embodiments, server 111 may include at least one storage node and multiple computing nodes. The storage node in each server may be a software module with storage-related functions, and the computing node may be hardware with computing capabilities such as CPU, GPU, FPGA, or virtual machine, container, process, or thread.

[0046] In some embodiments, at least one storage medium may be one or a combination of high-bandwidth memory (HBM), phase-change memory (PCM), resistive random access memory (RRAM), static random access memory (SRAM), dynamic random access memory (DRAM), magnetoresistive random access memory (MRAM), flash memory, 3D XPoint, and hard disk drive (HDD). The storage device may include various types of storage media to meet different data storage needs.

[0047] In some embodiments, storage network 112 may be a high-performance switch network. For example, storage network 112 may be any one or a combination of CXL (Compute Express Link), PCIe (PCI Express), SAS (Serial Attached SCSI), IB (InfiniBand), RoCE (RDMA over Converged Ethernet), and Ethernet networks. Compute nodes and storage nodes in different servers can access the shared storage pool through the switch storage network.

[0048] For example, using a SAS network as a data transmission channel offers advantages over traditional TCP / IP-based data storage methods. SAS transmission channels offer high performance, high bandwidth, low latency, and strong scalability. They can connect a large number of disk devices via SAS expanders, meeting the system's requirements for storing and transmitting large amounts of data and are more suitable for distributed architectures. After connecting to the SAS interface on a host adapter (HBA) or server motherboard, the storage provided by the SAS switch can be accessed simultaneously by multiple connected servers.

[0049] In practical applications, depending on the version and configuration of the SAS network, its supported bandwidth ranges from 3GB to 24GB, which is tens of times that of Gigabit Ethernet and several times that of the expensive 10 Gigabit Ethernet. Furthermore, at the transport layer, the "three-way handshake and four-way handshake" process required for data transmission based on the TCP protocol results in high communication overhead. In addition, the TCP protocol's delayed acknowledgment mechanism and slow start can lead to communication delays on the order of 100 microseconds. Compared to the TCP protocol, the SAS protocol's latency is only a fraction of that of the TCP protocol, significantly reducing communication latency.

[0050] In other words, different servers establish data transmission channels with the storage medium (shared storage pool) through a switch network, enabling data sharing between different servers. On the one hand, this saves the intermediate process of reading data into the memory of a specific server and then transmitting it to the memory of other servers via network protocols such as TCP / IP (horizontal data transmission process), reducing data transmission overhead. On the other hand, a high-performance switching network can provide high-throughput, low-latency data transmission, which can meet the system's needs for large-scale quantum state data storage and transmission.

[0051] Considering that in practical applications, external factors or internal system failures may cause the physical link between the storage node and the storage medium to be broken, affecting the server's access to the storage medium and thus impacting the performance of the entire system, in order to avoid the above situation, in some embodiments, each server and each storage medium in the system are connected to at least two networks. That is, there are at least two data transmission channels between each server and each storage medium, so that any server and any storage medium have two independent physical links. If any link is broken, the server can access the storage medium through the other link, giving the quantum simulation system good fault tolerance and thus ensuring the high availability of the system.

[0052] In some embodiments, servers are connected via a storage network.

[0053] In some embodiments, the storage network connecting the server can be one or any combination of Ethernet, IB, and RoCE.

[0054] In some embodiments, the shared storage pool is mounted on at least two servers as block devices. For example, protocols such as SCSI (Small Computer System Interface), iSCSI (Internet Small Computer System Interface), rbd (Rados Block Device), FC (Fiber Channel), NVME, NVMEoF (NVME over Fabrics), ATA, ATAPI, and FireWire can be used to mount the shared storage pool as one or more block devices on the at least two servers. For example, each server can mount one or more block devices, and each block device corresponds to a specific range of data in the shared storage pool. The quantum simulation software on each server can directly read the quantum state data as a block device, or it can create a file system on the block device to read the quantum state data as a file.

[0055] In some embodiments, the block device is a shared block device, that is, multiple servers share the same block device.

[0056] In some embodiments, the quantum simulation system may further include a management module (not shown) for managing devices or resources or centralization within the system. Specifically, it can manage and allocate shared storage pools, group quantum gates according to quantum circuits, and formulate a quantum simulation computation plan for each server based on each group of quantum gates, including the quantum simulation computation tasks required by the server when simulating that group of quantum gates.

[0057] In some embodiments, before the quantum simulation system performs simulation operations, the total amount of quantum state data space can be determined based on the properties of the quantum circuit (e.g., calculated based on the number of qubits to be simulated, whether it is full amplitude, partial amplitude, or tensor network, etc.), and the size of the shared storage pool can be determined based on the total amount of data.

[0058] Taking a 50-qubit full-amplitude quantum simulation system as an example, its total quantum state data space is 250 complex numbers. At a precision of complex128 (16-byte complex numbers), the total quantum state data space size is 250 × 16 = 16P, meaning at least 16P of shared storage is needed to store the quantum state data in the entire simulated quantum circuit. Specifically, the number of storage media required can be determined based on the size of the storage medium in a single storage device. For example, if each storage medium has a storage space of 16T, then at least 1024 storage media are needed.

[0059] In some embodiments, as shown in Figure 3, each server is responsible for simulating quantum state data within a corresponding range. For example, the range of state data that each server's computing node is responsible for is from 2n×index×Accuracy / C to 2n×(index+1)×Accuracy / C-1, where n is the number of qubits, index is the current computing node number (starting from 0), accuracy is the setting precision of the quantum state data (8 bytes for complex64 and 16 bytes for complex128), and C is the total number of computing nodes.

[0060] It should be noted that when simulating different quantum gates, the data range handled by each server may be different.

[0061] It should be noted that the number of servers, the number of storage media in the shared storage pool, the number of computing nodes in the servers, and the scope of each computing node in Figure 3 are merely examples and are not specifically limited.

[0062] Based on the aforementioned shared storage pool, the storage medium is decoupled from the server and does not belong to any single server. Instead, it is shared by all servers. Each server can directly access any storage medium in the shared storage pool through the storage network without relying on other servers. Based on this, the quantum simulation system can complete the entire quantum simulation calculation process through the collaborative work of multiple servers. In this process, there is no need for horizontal quantum state data transmission between different servers, eliminating the east-west data flow of the distributed quantum simulation cluster. This effectively solves the biggest bottleneck problem of distributed expansion of the quantum simulation system, improves the distributed expansion capability of the quantum simulation system, and is suitable for the transmission of large-scale quantum state data.

[0063] Based on the aforementioned quantum simulation system, this application further provides a quantum state data transmission method. In some embodiments, the aforementioned quantum state data transmission method can be integrated in the form of instructions into at least two servers 111 and at least one storage medium in the quantum simulation system 100, thereby enabling the servers to execute the quantum state data transmission method provided in this application.

[0064] Exemplary quantum state data transmission method

[0065] To further illustrate the transmission and storage process of quantum state data in the quantum simulation system shown in Figure 1 during simulated quantum computing tasks, this application also provides an exemplary flowchart (Figure 2) of a quantum state data transmission method. This quantum state data transmission method can be applied, for example, to the quantum simulation system 100 shown in the figure, or to other systems with corresponding processing capabilities.

[0066] In some embodiments, the process shown in FIG2 can be executed by a server in the quantum simulation system 100.

[0067] As shown in Figure 2, the method may include the following steps:

[0068] S210. Read some of the quantum state data required by the server to perform the simulated quantum computing task from the shared storage pool via the storage network.

[0069] S220. Based on this part of the quantum state data, perform simulated quantum computing tasks.

[0070] S230. Store the quantum state data generated after performing the simulated quantum computing task to the shared storage pool.

[0071] The partial quantum state data obtained in step S210 can be the partial quantum state data required for the simulated quantum computing task executed by the server, stored in the shared storage pool, i.e., the initial state quantum state data, the starting point of the simulated quantum computing task. After step S230, i.e., after executing the simulated quantum computing, the quantum state data includes the state data (intermediate data) generated throughout the simulation process and the final result quantum state data.

[0072] Simulated quantum computing tasks can refer to the process of simulating corresponding quantum computing on a server based on initial quantum state data.

[0073] In some embodiments, simulating a quantum computing task can be simulating at least one quantum gate operating on quantum state data. Specifically, it can simulate the state of a qubit and its quantum gate operations, or it can simulate a complex quantum algorithm, without being specifically limited to this.

[0074] A storage network can refer to a system that provides a data transmission channel between different target devices (such as storage media in servers and storage devices) by connecting them. A shared storage pool can consist of all or part of the storage media included in a quantum simulation system, enabling different servers to share quantum state data.

[0075] In some embodiments, each server reads some or all of the quantum state data required for the simulated quantum computing task performed by the server from a shared storage pool via a storage network, and stores the read quantum state data in the cache of the corresponding server. For example, it may first read a portion of the quantum state data related to the current sub-simulation task in the simulated quantum computing task, or it may read all the quantum state data required for the simulated quantum computing task. The specific determination can be made according to the server's cache, and no specific limitation is made here.

[0076] It should be noted that neither the data reading process in step S210 nor the data storage process in step S230 requires the transmission of quantum state data between different servers in at least two servers.

[0077] In some embodiments, the storage network can be a high-performance switch network, which, when combined with the aforementioned shared storage pool, can further reduce network latency, improve bandwidth utilization, and enhance data transmission performance. For details, please refer to the relevant description in Figure 1, which will not be repeated here.

[0078] In some embodiments, the server can send a request to the shared storage pool to read the quantum state data required for the server's simulated quantum computing task through the interface protocol of the storage medium in the shared storage pool. Based on this request, the server reads the quantum state data from the shared storage pool through the storage network. The interface protocol of the storage medium is the same as the protocol used by the storage network. That is, in the data reading process of step S210 and the data storage process of step S230, the interface protocol of the storage medium is the same as the protocol used by the storage network. In this case, the data path for the server to read and write to the shared storage pool is very similar to the data path for the server to read and write to the local disk. It is equivalent to extending the data cable from the motherboard to the local disk outside the server. Compared with reading and writing to the local disk, it only increases the loss through the switch in the storage network (since the data signal propagates at near the speed of light in the data cable, the performance loss increased by extending the data cable itself is negligible). For common protocols such as SAS / PCIe, the performance loss of the switch is very low, saving the protocol conversion process, reducing performance bottlenecks, and further improving system performance.

[0079] In some embodiments, the quantum computing simulated by the server is the quantum state data of the server within the corresponding range in the shared storage pool. That is, each server simulates the quantum state data within the corresponding range in the shared storage pool. There is no need to transmit quantum state data between different servers. For details about the corresponding range, please refer to the relevant description in Figure 1, which will not be repeated here.

[0080] Specifically, a simulation task plan can be formulated for each server based on the type and parameters of at least one quantum gate. The simulation task plan can be understood as a set of quantum simulation computation tasks that the server needs to perform.

[0081] Considering that in some embodiments, some servers may perform simulated quantum computing tasks while others are idle, in some embodiments, the simulation task plan may include zero or at least one simulation operation, wherein at least one simulation operation includes the quantum state data range corresponding to the simulation operation and the quantum computing operation to be simulated (i.e., simulated quantum computing task). Based on the simulation task plan, each server completes the simulated computation of the specified quantum state data.

[0082] In some embodiments, to address the high data throughput requirements of large-scale quantum simulation systems, at least two servers and at least one storage medium are connected via a storage network. This establishes a connection between each server and each storage medium, constructing at least one storage medium as a shared storage pool, thus decoupling the storage medium from the servers. In other words, all storage media function as a single, integrated storage unit, not belonging to any single server; all servers share the storage medium. Based on this, the entire quantum simulation computation process can be completed without lateral quantum state data transfer between different servers, eliminating east-west data flow in the distributed quantum simulation cluster. This effectively solves the biggest bottleneck problem in the distributed scaling of quantum simulation computation, improves the distributed scalability of the quantum simulation system, and further enhances the overall system performance.

[0083] In addition, based on the shared storage pool, multiple servers can work together to complete the overall quantum computing task of the quantum simulation system. That is, each server simulates quantum state data within its corresponding range. On the one hand, this can improve the processing efficiency of the quantum simulation system. On the other hand, each server simulating the corresponding quantum state data independently can reduce the computational complexity, which is beneficial to improving the accuracy and stability of the simulation results.

[0084] In some embodiments, after the server executes the corresponding simulated quantum computing task, the quantum state data after the simulated quantum computing task is stored in a shared storage pool.

[0085] Therefore, the aforementioned quantum state data transmission method establishes a connection between each server and each storage medium through a storage network, constructing all storage media into a shared storage pool. The quantum state data is stored in the shared storage pool independent of each server, rather than in the private storage media of each server. This allows any server to directly access any storage medium through the storage network without relying on other servers during the entire quantum simulation process. This eliminates the east-west data flow of the distributed quantum simulation cluster, effectively solving the biggest bottleneck problem in the distributed expansion of quantum simulation computing, improving the distributed expansion capability of the quantum simulation system, and further improving the overall performance of the system to meet the data transmission requirements of large-scale quantum simulation systems.

[0086] Exemplary device

[0087] The method and system embodiments of this application have been described in detail above with reference to Figures 1 to 3. The device embodiments of this application are described in detail below. It should be understood that the descriptions of the method embodiments correspond to the descriptions of the device embodiments; therefore, any parts not described in detail can be referred to the preceding method embodiments. Figure 4 is a schematic block diagram of a quantum state data transmission device shown in some embodiments of this application.

[0088] As shown in Figure 4, the quantum state data transmission device 400 may include a data reading module 410, an execution module 420, and a data storage module 430.

[0089] The data reading module 410 can be configured to read partial quantum state data required by the server to simulate quantum computing tasks from a shared storage pool via a storage network.

[0090] The execution module 420 can be configured to perform simulated quantum computing tasks based on this portion of quantum state data.

[0091] The data storage module 430 can be configured to store the quantum state data after performing a simulated quantum computing task to a shared storage pool.

[0092] It should be understood that specific limitations regarding quantum state data transmission devices can be found in the limitations regarding quantum state data transmission methods described above, and will not be repeated here. Each module in the aforementioned quantum state data transmission device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in hardware or independently of the processor in a computer device, or stored in software in the memory of a computer device, so that the processor can call and execute the operations corresponding to each module.

[0093] Exemplary electronic devices and storage media

[0094] This application also provides an electronic device, as shown in FIG5. The electronic device 500 provided by this application may include a memory 510, a processor 520, and an input / output interface 530. The memory 510, processor 520, and input / output interface 530 are connected via internal connection paths. The memory 510 is used to store instructions, and the processor 520 is used to execute the instructions stored in the memory 510 to control the input / output interface 530 to receive input data and information, and output operation results and other data.

[0095] It should be understood that in the embodiments of this application, the processor 520 may be a general-purpose central processing unit (CPU), GPU, FPGA, microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits to execute related programs in order to implement the technical solutions provided in the embodiments of this application.

[0096] The memory 510 may include read-only memory and random access memory, and provides instructions and data to the processor 520. A portion of the processor 520 may also include non-volatile random access memory. For example, the processor 520 may also store device type information.

[0097] In implementation, each step of the above method can be completed by the integrated logic circuits in the hardware of the processor 520 or by instructions in software form. The quantum state data transmission method disclosed in the embodiments of this application can be directly implemented by the hardware processor, or by a combination of hardware and software modules in the processor. The software modules can be located in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. This storage medium is located in memory 510, and the processor 520 reads the information in memory 510 and completes the steps of the above method in combination with its hardware. To avoid repetition, it will not be described in detail here. This application also provides a computer program product, including a computer program / instructions. When the computer program / instruction processor in the computer program product provided in this application is executed, the quantum state data transmission method provided in this application can be implemented.

[0098] All of the above-mentioned optional technical solutions can be combined in any way to form the optional embodiments of this application, and will not be described in detail here.

[0099] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0100] It should be noted that in the apparatus, devices, and methods of this application, each module or step can be disassembled and / or recombined. These disassemblies and / or recombinations should be considered equivalent solutions of this application. The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use this application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein can be applied to other aspects without departing from the scope of this application. Therefore, this application is not intended to be limited to the foregoing aspects, but rather to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0101] The above description is for illustrative purposes only, illustrating and describing the technical solutions of this application. Furthermore, this description is not intended to limit the embodiments of this application to the scope of the forms disclosed above. Although several exemplary aspects and embodiments have been discussed above, those skilled in the art can readily derive other variations, modifications, alterations, additions, and sub-combinations based on the above content.

[0102] It should be noted that in the description of this application, the terms "first," "second," "third," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance. Furthermore, in the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0103] The above are merely preferred embodiments of this application and are not intended to limit this application. Any modifications or equivalent substitutions made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method of quantum state data transmission, characterized by, The method is applied to a quantum simulation system, which includes at least two servers, at least one storage medium, and a storage network. The at least one storage medium constitutes a shared storage pool for storing quantum state data. Either of the at least two servers can access any storage medium in the shared storage pool through the storage network without going through other servers. The method includes: The storage network is used to read partial quantum state data required by the server to perform simulated quantum computing tasks from the shared storage pool. Based on the aforementioned partial quantum state data, the simulated quantum computing task is executed; The quantum state data generated after performing the simulated quantum computing task is stored in the shared storage pool.

2. The quantum state data transmission method of claim 1, wherein, The at least one storage medium is located in at least one storage device independent of the at least two servers, and the at least one storage medium is separate from the at least two servers.

3. The quantum state data transmission method of claim 1, wherein, The shared storage pool is mounted on the at least two servers as a block device.

4. The quantum state data transmission method of claim 3, wherein, The block device is a shared block device.

5. The quantum state data transmission method according to any one of claims 1 to 4, characterized in that, The step of reading the partial quantum state data required for the server to simulate the quantum computing task from the shared storage pool via the storage network includes: The quantum state data can be read from the shared storage pool via the storage network, either partially or entirely. Some or all of the quantum state data are stored in the cache of the server.

6. The quantum state data transmission method according to any one of claims 1 to 5, wherein, During the execution of the steps of reading partial quantum state data required for the simulated quantum computing task from the shared storage pool through the storage network and storing the quantum state data generated after executing the simulated quantum computing task into the shared storage pool, the interface protocol of the storage medium is the same as the protocol used by the storage network.

7. The quantum state data transmission method according to any one of claims 1 to 6, characterized by, The storage network is any one or a combination of CXL, PCIe, SAS, IB, RoCE, and Ethernet networks.

8. The quantum state data transmission method according to any one of claims 1 to 7, characterized by, The at least one storage medium is one or a combination of high-bandwidth memory (HBM), phase-change memory (PCM), resistive random access memory (RRAM), static random access memory (SRAM), dynamic random access memory (DRAM), magnetoresistive random access memory (MRAM), flash memory, 3D XPoint, and hard disk drive (HDD).

9. The quantum state data transmission method according to any one of claims 1 to 8, wherein, The simulated quantum computing task includes simulating the operation of at least one quantum gate on the quantum state data, wherein the simulated quantum computing task performs specified simulated quantum computing operations on a specified range of quantum state data.

10. The quantum state data transmission method of claim 9, wherein, The simulation of at least one quantum gate operating on the quantum state data includes: Based on the type and parameters of the at least one quantum gate, a set of simulated quantum computing tasks is determined for each of the at least two servers, wherein the set of simulated quantum computing tasks consists of zero or at least one simulated quantum computing task.

11. A quantum simulation system, characterized by, The system comprises at least two servers, at least one storage medium, and a storage network, wherein the at least one storage medium constitutes a shared storage pool for storing quantum state data, any server of the at least two servers accesses any storage medium in the shared storage pool through the storage network, and the at least two servers are configured to perform the quantum state data transmission method of any one of claims 1-10.

12. A quantum state data transmission apparatus, characterized by comprising: The device is applied to a quantum simulation system comprising at least two servers, at least one storage medium, and a storage network, wherein the at least one storage medium constitutes a shared storage pool, any server of the at least two servers accesses any storage medium in the shared storage pool through the storage network without passing through other servers, and the device comprises: a data reading module configured to read, through the storage network, partial quantum state data required by the server for performing a simulated quantum computing task from the shared storage pool; a task execution module configured to perform the simulated quantum computing task based on the partial quantum state data; and a data storage module configured to store quantum state data generated after performing the simulated quantum computing task to the shared storage pool.

13. An electronic device, comprising: The electronic device comprises: at least one processor; and a memory connected to the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the quantum state data transmission method of any one of claims 1-10.

14. A computer-readable storage medium, characterized in that, The computer readable storage medium stores instructions, and when the instructions in the computer readable storage medium are executed by the processor of the electronic device, the electronic device can perform the quantum state data transmission method of any one of claims 1-10.